Clinical problems exist as to how much cortisol should optimally be given to patients who are deficient in hydrocortisone, either because of pituitary ACTH failure or adrenal failure. This protocol examines the hypothesis that detailed study of ACTH and cortisol, and urine free cortisol in patients receiving replacement hydrocortisone therapy may allow the more rational and accurate selection of replacement therapy.